RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation

Tianjiao Li, Mengran Yu, Chenyu Shi, Yanjun Zhao, Xiaojing Liu, Qi Zhang, Xuanjing Huang, Qiang Zhang, Jiayin Wang


Abstract
Large language models (LLMs) possess strong multilingual capabilities, and combining Reinforcement Learning from Human Feedback (RLHF) with translation tasks has shown great potential. However, we observe that this paradigm performs unexpectedly poorly when applied to colloquial subtitle translation tasks. In this work, we investigate this issue and find that the offline reward model (RM) gradually diverges from the online LLM due to distributional shift, ultimately leading to undesirable training outcomes. To address this, we propose RIVAL, an adversarial training framework that formulates the process as a min–max game between the RM and the LLM. RIVAL iteratively updates the both models, with the RM trained to distinguish strong from weak translations (qualitative preference reward), and the LLM trained to enhance its translation for closing this gap. To stabilize training and improve generalizability, we also incorporate quantitative preference reward (e.g., BLEU) into the RM, enabling reference-free quality modeling aligned with human evaluation. Through extensive experiments, we demonstrate that the proposed training framework significantly improves upon translation baselines.
Anthology ID:
2025.findings-emnlp.166
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2025
Month:
November
Year:
2025
Address:
Suzhou, China
Editors:
Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3064–3079
Language:
URL:
https://aclanthology.org/2025.findings-emnlp.166/
DOI:
Bibkey:
Cite (ACL):
Tianjiao Li, Mengran Yu, Chenyu Shi, Yanjun Zhao, Xiaojing Liu, Qi Zhang, Xuanjing Huang, Qiang Zhang, and Jiayin Wang. 2025. RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 3064–3079, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation (Li et al., Findings 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.findings-emnlp.166.pdf
Checklist:
 2025.findings-emnlp.166.checklist.pdf